The first two years of life are increasingly recognized as a critical period for the development of obesity. Prior studies demonstrate the importance of parents in shaping this risk with particular emphasis on associations between maternal responsive feeding (i.e., higher responsiveness to child hunger and satiety cues and fewer assertive prompts to eat) and lower child BMI. However, research to date has been primarily cross-sectional, based on parent-report, and done almost exclusively in preschoolers or older children; longitudinal studies of mothers and toddlers that assess objective measures of responsive feeding within the context of antecedent and concurrent factors, are needed to better understand factors that promote responsive feeding and reduce long-term child obesity risk. We propose to enroll 164 mother-toddler dyads into a prospective study of feeding interactions and obesity risk, with assessments at 18, 24, and 36 months, in order to achieve the following aims: 1) understand factors related prospectively to responsive feeding between mothers and toddlers at 18 months; 2) determine whether responsive feeding at 18 and 24 months predicts change in zBMI and adiposity from 18-36 months; and 3) explore which factors are most strongly predictive of child zBMI change from birth to 36 months. Responsive feeding will be assessed with two home-based meal observations at both 18 and 24 months (total of 4). At these time points, mothers will also complete three 24-hour dietary recalls of their own and their toddler's intake, and a battery of measures assessing additional maternal (e.g., perceptions of feeding practices and styles, dietary restraint) and toddler (e.g., temperament, appetite) factors that could influence maternal responsiveness and toddler obesity risk. Mothers' and toddlers' heights, weights and circumference/skinfolds will be measured at 18, 24, and 36 months. We propose to recruit participants from Healthy Beginnings (HB), an NHLBI-funded, two site randomized controlled trial testing the efficacy of a multi- component behavioral intervention in reducing risk of excessive weight gain during pregnancy. HB provides a unique opportunity to cost-effectively conduct this longitudinal study: participants are already being enrolled in HB; al are overweight or obese and half are Latina (important risk factors for child obesity); relevant assessments of women in HB occur at study entry (<16 weeks pregnancy), 22-24 weeks and 35 weeks of pregnancy, and both mothers and babies at delivery, 6 and 12 months. Half of HB participants were randomized to an intervention designed to reduce pregnancy weight gain, providing a unique opportunity to examine long-term effects on subsequent child obesity risk. HB ends when infants are 12 months old. Extending follow-up through 36 months will enable understanding of longitudinal associations between early life influences on responsive feeding during toddlerhood, and prospective associations between responsive feeding and obesity risk. Findings will have important implications for preventive interventions during this important and understudied developmental period.